swamirishi commented on code in PR #8260: URL: https://github.com/apache/ozone/pull/8260#discussion_r2040114943
########## hadoop-ozone/ozone-manager/src/main/java/org/apache/hadoop/ozone/om/service/CompactionService.java: ########## @@ -0,0 +1,159 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.hadoop.ozone.om.service; + +import com.google.common.annotations.VisibleForTesting; +import java.io.IOException; +import java.util.List; +import java.util.concurrent.TimeUnit; +import java.util.concurrent.atomic.AtomicBoolean; +import java.util.concurrent.atomic.AtomicLong; +import org.apache.hadoop.hdds.utils.BackgroundService; +import org.apache.hadoop.hdds.utils.BackgroundTask; +import org.apache.hadoop.hdds.utils.BackgroundTaskQueue; +import org.apache.hadoop.hdds.utils.BackgroundTaskResult; +import org.apache.hadoop.hdds.utils.db.RDBStore; +import org.apache.hadoop.hdds.utils.db.RocksDatabase; +import org.apache.hadoop.hdds.utils.db.managed.ManagedCompactRangeOptions; +import org.apache.hadoop.ozone.om.OMMetadataManager; +import org.apache.hadoop.ozone.om.OzoneManager; +import org.apache.hadoop.util.Time; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +/** + * This is the background service to compact OM rocksdb tables. + */ +public class CompactionService extends BackgroundService { + private static final Logger LOG = + LoggerFactory.getLogger(CompactionService.class); + + // Use only a single thread for Compaction. + private static final int COMPACTOR_THREAD_POOL_SIZE = 1; + + private final OzoneManager ozoneManager; + private final OMMetadataManager omMetadataManager; + private final AtomicLong numCompactions; + private final AtomicBoolean suspended; + // list of tables that can be compacted + private final List<String> compactableTables; + + public CompactionService(OzoneManager ozoneManager, TimeUnit unit, long interval, long timeout, + List<String> tables) { + super("CompactionService", interval, unit, + COMPACTOR_THREAD_POOL_SIZE, timeout, + ozoneManager.getThreadNamePrefix()); + this.ozoneManager = ozoneManager; + this.omMetadataManager = this.ozoneManager.getMetadataManager(); + + this.numCompactions = new AtomicLong(0); + this.suspended = new AtomicBoolean(false); + for (String table : tables) { + if (!omMetadataManager.listTableNames().contains(table)) { + throw new IllegalArgumentException(); + } + } + this.compactableTables = tables; + } + + /** + * Suspend the service (for testing). + */ + @VisibleForTesting + public void suspend() { + suspended.set(true); + } + + /** + * Resume the service if suspended (for testing). + */ + @VisibleForTesting + public void resume() { + suspended.set(false); + } + + /** + * Returns the number of manual compactions performed. + * + * @return long count. + */ + @VisibleForTesting + public long getNumCompactions() { + return numCompactions.get(); + } + + @Override + public synchronized BackgroundTaskQueue getTasks() { + BackgroundTaskQueue queue = new BackgroundTaskQueue(); + for (String tableName : compactableTables) { + queue.add(new CompactTask(tableName)); + } + return queue; + } + + private boolean shouldRun() { + return !suspended.get(); + } + + protected void compactFully(String tableName) throws IOException { + long startTime = Time.monotonicNow(); + LOG.info("Compacting column family: {}", tableName); + try (ManagedCompactRangeOptions options = new ManagedCompactRangeOptions()) { + options.setBottommostLevelCompaction(ManagedCompactRangeOptions.BottommostLevelCompaction.kForce); + RocksDatabase rocksDatabase = ((RDBStore) omMetadataManager.getStore()).getDb(); + + try { + // Find CF Handler + RocksDatabase.ColumnFamily columnFamily = rocksDatabase.getColumnFamily(tableName); + rocksDatabase.compactRange(columnFamily, null, null, options); Review Comment: Can we only do it based on some sst file metadata. Based on some estimate of overlapping key ranges in the sst file and based on these thresholds. It doesn't really make sense to compact to bottom most blindly. This can have a huge impact on ozone snapshot scale till we work snapshot compactions. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
